Run mcRigor metacell partition assessment
Usage
RunmcRigor(
srt,
cell_membership = NULL,
metacell.by = NULL,
mode = c("detect", "optimize"),
tgamma = NULL,
gamma_names = NULL,
assay_type = c("RNA", "ATAC"),
Gammas = NULL,
aggregate_method = c("mean", "sum", "geom"),
output_file = NULL,
Nrep = 1,
gene_filter = 0.1,
feature_use = 2000,
cor_method = c("pearson", "spearman"),
prePro = TRUE,
test_cutoff = 0.01,
thre_smooth = TRUE,
thre_bw = 1/6,
D_bw = 10,
optim_method = c("tradeoff", "dub_rate_large", "dub_rate_small"),
weight = 0.5,
dub_rate = 0.1,
draw = FALSE,
pur_metric = NULL,
check_purity = TRUE,
fields = NULL,
step_save = FALSE,
prefix = "mcRigor",
tool_name = "mcRigor",
verbose = TRUE
)Arguments
- srt
A
Seuratobject containing the original single-cell data.- cell_membership
A data frame or matrix with cells in rows and one or more metacell partitions in columns. Row names should be cell names. If row names are missing and the row count equals
ncol(srt), cells are matched incolnames(srt)order.- metacell.by
Metadata column(s) in
srtused as metacell partitions whencell_membership = NULL.- mode
McRigor task.
"detect"callsmcRigor_DETECT()for one partition;"optimize"callsmcRigor_OPTIMIZE()across candidate partitions.- tgamma
Target partition/gamma for
"detect". Can be a membership column name or the numeric gamma label used by mcRigor. IfNULL, the first membership column is used.- gamma_names
Optional gamma labels for membership columns. mcRigor requires numeric-like column labels; non-numeric labels are mapped internally to
1:ncol(cell_membership)and recorded in the stored result.- assay_type
Assay type passed to mcRigor.
- Gammas
Candidate gamma labels for
"optimize". Can use original membership column names or mapped mcRigor gamma labels.- aggregate_method
Metacell aggregation method passed to mcRigor.
- output_file
Optional path where mcRigor writes the
TabMCRDS file. IfNULL, a temporary file is used to avoid creating files in the working directory.- Nrep
Number of permutation repetitions used by mcRigor.
- gene_filter, feature_use, cor_method, prePro, test_cutoff, thre_smooth, thre_bw
Parameters forwarded to mcRigor.
- D_bw, optim_method, weight, dub_rate
Optimization parameters forwarded to
mcRigor_OPTIMIZE().- draw, pur_metric, check_purity, fields, step_save
Plotting, purity, and intermediate-save parameters forwarded to mcRigor.
- prefix
Prefix for metadata columns written to
srt.- tool_name
Name of the
srt@toolsentry used to store results.- verbose
Whether to print the message. Default is
TRUE.
References
Liu, P. and Li, J.J. (2024). mcRigor: a statistical method to enhance the rigor of metacell partitioning in single-cell data analysis. bioRxiv. doi:10.1101/2024.10.30.621093
Examples
data(pancreas_sub)
set.seed(11)
pancreas_sub <- RunStandardWorkflow(
pancreas_sub,
nHVF = 500,
linear_reduction_dims = 20,
linear_reduction_dims_use = 1:20,
nonlinear_reduction_dims = 2,
verbose = FALSE
)
#> ℹ [2026-08-30 05:52:57] Skip `log1p()` because `layer = data` is not "counts"
mc <- RunMetaCell(
pancreas_sub,
method = "supercell",
gamma = 25
)
#> ℹ [2026-08-30 05:53:07] Running SuperCell with gamma = 25, k.knn = 5 on 1000 cells
#> ℹ [2026-08-30 05:53:08] `RunMetaCell()` ("supercell") built 40 metacells from 1000 cells
#> ℹ [2026-08-30 05:53:08] Metacell size summary: min 5, median 20, mean 25, max 74 cells
#> Warning: Data is of class dgeMatrix. Coercing to dgCMatrix.
#> ✔ [2026-08-30 05:53:08] `RunMetaCell()` returned metacell Seurat with 40 metacells. Original cells in `@misc[["original_srt"]]`
membership <- data.frame(
Metacell = mc@misc[["cell_membership"]],
row.names = names(mc@misc[["cell_membership"]])
)
pancreas_sub <- RunmcRigor(
mc@misc[["original_srt"]],
cell_membership = membership,
Nrep = 1,
feature_use = 100,
draw = FALSE
)
#> ℹ [2026-08-30 05:53:47] Run mcRigor in "detect" mode with 1 partition
#> Normalizing data...
#> gamma = 1
#>
|
| | 0%
|
|== | 2%
|
|==== | 5%
|
|===== | 8%
|
|======= | 10%
|
|========= | 12%
|
|========== | 15%
|
|============ | 18%
|
|============== | 20%
|
|================ | 22%
|
|================== | 25%
|
|=================== | 28%
|
|===================== | 30%
|
|======================= | 32%
|
|======================== | 35%
|
|========================== | 38%
|
|============================ | 40%
|
|============================== | 42%
|
|================================ | 45%
|
|================================= | 48%
|
|=================================== | 50%
|
|===================================== | 52%
|
|====================================== | 55%
|
|======================================== | 57%
|
|========================================== | 60%
|
|============================================ | 62%
|
|============================================== | 65%
|
|=============================================== | 68%
|
|================================================= | 70%
|
|=================================================== | 72%
|
|==================================================== | 75%
|
|====================================================== | 78%
|
|======================================================== | 80%
|
|========================================================== | 82%
|
|============================================================ | 85%
|
|============================================================= | 88%
|
|=============================================================== | 90%
|
|================================================================= | 92%
|
|================================================================== | 95%
|
|==================================================================== | 98%
|
|======================================================================| 100%
#>
#> dubious trustworthy
#> 27 13
#> ✔ [2026-08-30 05:53:53] mcRigor results stored in `srt@tools[[mcRigor]]`
table(pancreas_sub$mcRigor_status)
#>
#> dubious trustworthy
#> 754 246
CellDimPlot(
pancreas_sub,
group.by = "mcRigor_metacell"
)
CellDimPlot(
pancreas_sub,
group.by = "mcRigor_status"
)
